This manuscript has been reproduced from the microfilm master. UMI films
the text directly from the original or copy submitted. Thus, som e thesis and
dissertation copies are In typewriter ^ c e , while others may be from any type of
computer printer.
The quality of th is reproduction Is dependent upon th e quality of th e
copy subm itted. Broken or Indistinct print, colored or poor quality Illustrations
and photographs, print bleedthrough, substandard margins, and Improper
alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete manuscript
and there are missing pages, these will be noted. Also, If unauthorized
copyright material had to be removed, a note will Indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand comer and continuing
from left to right In equal sections with small overlaps.
Photographs Included In the original manuscript have been reproduced
xerographlcally In this copy.
Higher quality 6" x 9” black and white
photographic prints are available for any photographs or Illustrations appearing
In this copy for an additional charge. Contact UMI directly to order.
Bell & Howell Information and Learning
300 North Zeeb Road, Ann Ariaor, Ml 48106-1346 USA
800-521-0000
by
Sheila Chiwocha
B.Sc., University of Guelph, 1994
A D issertation Subm itted in Partial Fulfillment of th e R equirem ents for the
D egree of
DOCTOR OF PHILOSOPHY
in the Departm ent of Biology
W e a c c e p t this dissertation a s conforming to the required standard
Dr. P. von A derkas, Supervisor (D epartm ent of Biology)
Dr. B.d. Hawkins, M em ber (D epartm ent of Biology)
Dr. L. P ag e, M em ber (D epartm ent of Biology)
son. O utside M ember (D epartm ent of Biochemistry and Microbiology)
Dr. D. Cyr, External Examiner (Cellfor Inc.)
© Sheila Chiwocha
University of Victoria
All rights reserved. This dissertation m ay not b e reproduced in w hole or in part,
by photocopying or other m eans, without the perm ission of the author.
ABSTR.ACT
M cgagam etophyie developm ent in sim and in vitro was investigated in Douglas fir to address the following questions: 1) Do endogenous levels o f plant horm ones change during m egagam etoph\'te developm ent and arc they associated with m orphological changes? 2) Can mcgagametopliytes be cultured prior to fertilization? 3) Can em bryos be rescued from megagametoph) tes cultured soon alter fertilization?
A histochemical study o f storage reserve deposition during m egagam etophyte developm ent was pcrfonued with material isolated weekly for 11 weeks. Prior to
fertilization, starch was detected in the neck cells o f m egagam etophytes analyzed 9 weeks alter pollination ( WAP». During em br.ogcnesis. starch was deposited in the central region o f megagametophytes. Proteins and lipids were llrst detected in the prothallial cells in the peripheiy o f megagametophytes isolated 14 and 15 W.-\P. respectively. With further developm ent, starch was deposited in prothallial cells around the corrosion cavity, w hile proteins and lipids were spatially localized to prothallial cells in the periphery. In the em bryo, starch accum ulation was preferentially localized in the root cap and the em bryonal suspensor cells at 17 W.AP.
A parallel study quantifying the endogenous levels o f plant hormones: lAA. lA A sp, Z, ZR. iP. IP .\. .AB.A. and AB.‘\-G E . in megagametophytes was performed. H orm ones were extracted, purified and fractionated using HPLC. To correct for losses due to procedures, radiolabelled standards were added prior to extraction. The hormones were quantified using an ELIS.A. method. On a dry weight basis. Z levels were highest in megagametophy'tes at the late central cell stage (8 WAP). During em bryogenesis. Z levels peaked during week 13. ZR peaked twice at 13 and 17 WAP. The iP content o f
m egagam etophytes increased at 10, 13 and 17 W.AP while iP.A concentration increased at 13 and 17 WAP. Prior to fertilization, the free LAA was highest in m egagam etophytes at 9 W AP. D uring embryogenesis, the m ajo rlA A accumulations occurred at I I , 13 and 15 WAP. lA A sp concentrations reached their highest levels at 10, 14 and 18 W AP. ABA
constant over the 11 weeks analyzed.
M egagam etophytes were isolated weekly from 7-10 W.AP and cultured on a modified half-strength Litvay's medium supplemented w ith one o f three auxins (N.AA. IBA or 2.4-D) and a cytokinin (2 m g'L B.AP). Each auxin was tested at three levels; 0.1.
1.0 or 10 m g'L. The objective was to determine w hether the megagametophytes would continue to grow in culture. ,\legagantetophytes increased in length after 9 and IS days o f culture. .Auxin and cytokinin supplements had a significant et'fect on growth for material isolated 7 or 10 W.AP. However, the viability o f the archegonia r a p id ly declined on all the m edia tested. The most optimal treatment for each auxin type ( B.AP in com bination with 0.1 m g'L N.A.A. 1.0 m g'L IB.A or 1.0 m g L 2.4-D) was used to initially culture pollinated megagam etophytes in the embryo rescue experiment.. .After 2 1 days, megagametophytes were transferred to media containing .AB.A concentrations o f 0. 5. 20 or 40 p.VL .A majority o f the rescued embryos were developmentally arrested at the globular stage. Only tiaree embryos, containing over 30 cotyledons each, matured on .AB.A concentrations o f 5. 20 or 40 uM.
In conclusion, the prothallial cells o f the pre-fertilization m egagam etophytes could be cultured for long periods and their growth was not dependent on the presence o f viable archegonia. The endogenous levels o f plant hormones varied w ith megagametophyte developm ent and were associated with morphological changes. This information has im plications for growing megagametophytes for in viiro fertilization and embryo rescue experim ents. The endogenous levels o f plant hormones could be used to design culture m edia for rescuing embrvos resulting from in viiro fertilization in Douglas fir.
Examiners:
Dr. P. von A derkas, Supervisor (D epartm ent of Biology)
Dr. B.J. Hawkins, Member (D epartm ent of Biology)
Dr. L. P ag e, M ember (Department of Biology)
n. Outside Member (D epartm ent of Biochemistry and Microbiology)
A bstract ... T ab le of C o n te n ts ... List of T ables ... List of F igures ... A cknow ledgm ents ... C h a p te r I: In tro d u c tio n ... C h a p te r 2: L ite r a tu r e review ... 2.1 C onifer .seed developm ent ...
2.1.1 L n i b r \ u g e n y ...
2 .1.2 Fertilization and zygotic embryogenesis ... 2.1.3 Storage reserves during conifer seed development 2.2 In vitro megagametophyte culture studies ... 2.3 Kmbryo rescue ... 2.3.1 Som atic embryogenesis ... 2.4 Plant hormones during seed developm ent ... 2.4.1 Background ... 2.4.2 A ngiosperm seed developm ent ... 2.4.2.1 The role o f ABA during angiosperm seed developm ent 2.4.2.1.1 Precocious germ ination ... 2.4.2.1.2 Storaue reserve accu m u latio n ...
X I X I V 15 18 19 21 21 n 23 24
2.4.2.1.4 Dormancy ... 32
2.4.2.2 O ther hormones during fruit and seed developm ent . 35 2.4.3 Plant hoim ones during embryogenesis in conifers . 37 C h a p te r 3: D ouglas fir m egagam etophyte dev elop m en t ... 40
3.1 Introduction ... 40
3.2 M aterials and .Methods ... 41
3.2.1 Plant M aterials ... 41
3.2.2 .Anatomical analysis o f megagametophyte developm ent 42 3.2.3 Histochem ical analysis o f megagametophyte development 42 3.3 Results ... 43
3.3.1 A natom ical analysis o f megagametophyte developm ent 43 3.3.2 Histochemical analysis of megagametophyte development 46 3.4 Discussion ... 51
C h a p te r 4: In vitro c u ltu re o f Douglas fir m eg agam etophytes ... 56
4.1 Introduction ... 56
4.2 M aterials and M ethods ... 60
4.2.1 Plant M aterials ... 60
4.2.1.1 Pre-fertilization megagametophytes ... 60
analysis ... 61
4.2.2.1 Pre-tertilization m egagamecoplmes ... 61
4.2.2.1.1 M edia com position ... 61
4.2.2.1.2 Experimental desian and statistical analysis ... 62
4.2.2.2 Post-fertilization megagam etopiu tes ... 63
4.2.2.2.1 M edia com position ... 63
4.2.2.2.2 Experimental design ... 64
4.3 R esults... 65
4.3.1 Pre-fcnilization megagametophytes ... 65
4.3.1.1 Length responses on the different auxin concentrations.... 65
4.3.1.2 Viability o f megagametophytes on \ arious media ... 78
4.3.2 Post-fertilization m eg agam etophytes... 88
4.4 Discussion ... 93
C h a p t e r s : E n d o g en o u s levels of p la n t horm ones d u rin g D ouglas fir m eg ag am etoph yte developm ent ... 97
5.1 Introduction ... 100
5.2 M aterials and M ethods ... 100
5.2.1 Plant M aterials ... 101
5.2.2 H ormonal analysis ... 101
5.2.2.1 Extraction ... 102
5.2.2.4 Q uantification o f plant hom iones ... 103
5.2.2.5 Determ ination o f correction factors for losses due to the procedures ... 104
5.1 Resiiit'; 106 5.3.1 M egagam elophvte fresh and diy weights ... 106
5.3.2 Hormonal analysis ... 106
5.3.2.1 Endogenous cytokinin concentration ... 108
5.3.2.2 Endogenous l.-\.\ and I .A. .Asp concentrations ... 111
5.3.2.3 Endogenous .AB.A and .AB.A-GE concentrations ... 115
5.4 Discussion ... 1 19 C h a p te r 6. G e n e ra l discussion ... 124
6.1 Question I. Do plant hormone levels change during megagametophyte development? ... 124
6. l.l Comparison o f plant hormone levels in this study with those from other conifer studies ... 128
6.2 Question 2. Are morphological and physiological changes during developm ent independent o f one another? ... 130
6.3 Question 3. C an m egagametophytes be cultured prior to fertilization? ... 132
6.4 Question 4. Can em bryos be rescued from megagametophytes cultured soon after fertilization? ... 132
T a b le I Tim e scale of developmental events in prc-lertilization and postfertilization Douglas fir m egagam etoph\ tes as a function
n f W A P... :2
T a b le 2 Statistical analysis o f lengths o f pre-fertilization Douglas fir megagametophytes after ^ days on
various culture m e d ia ... 71 T a b le 3 Statistical analysis o f viability o f pre-fertilization
Douglas fir megagametophytes at'ier 9 days on \arious
culture m edia... 83 T ab le 4 The num ber o f embia os rescued from Douglas t'lr
megagametophytes cultured 10-12 weeks after
pollination (WAP)... 92 T ab le 5 Plant hormones detected at high levels during the
different stages of Douglas fir megagametophyte
developm ent... 126 T a b le 6 A com parison of the levels o f plant hormones in seeds
o f different conifers studied to date... 129 T a b le 7 A com parison o f cytokinin levels in Douglas fir seeds
Figure la -d Sum m ar\' diagram o f the stages o f Douglas fir
m egagametophyte developm ent... 7
Figure 2 a -f Plate I... ... 15
Figure 3 a -f Plate 2. ... 48
Figure 4 a -f Plate 3. ... 50
Figure 5a A verage lengths of pre-fertilization megagametophytes at 0. 9 and 18 days in culture as a function o f W AP 66 Figure 5b A verage lengths o f post-prefertilization megagametophytes at 0, 9 and 18 days in culture as a function o f W AP 67 Figure 6a-cl Lengths o f megagametophytes isolated 7 WAP and cultured on m edia containing various combinations o f auxins and cytokinin... 69
Figure 7a-d Lengths o f megagametophytes isolated 8 WAP and cultured on m edia containing various combinations o f auxins and cytokinin... 73
Figure 8a-d Lengths o f megagametophytes isolated 9 WAP and cultured on m edia containing various combinations o f auxins and cytokinin... 75 F igure 9a-d Lengths o f megagametophytes isolated 10 W AP and cultured
F ig u re lO a-b Plate 4 80 F ig u re l l a - d Viability o f megagametophytes isolated 7 WAP and culuirec
on m edia containing \arious com binations o f auxins and
F ig u re I2 a -d Viability o f megagametophvies isolated 8 WAP and cultured on m edia containing \ arious com binations o f auxins and cvtokinin... 85 F ig u re 13a-d Viability o f megagametophytes isolated 9 WAP and cultured
on media containing various com binations o f auxins and cvtokinin... 87 F ig u re I4 a -h Plate 5 ...
F ig u re 15 Changes in fresh and dr}- weights o f megagametophytes over tim e... F ig u re 16 Endogenous levels o f Z and ZR. on a drv weight basis.
F ig u re 17
F ig u re 18
F ig u re 19
during Douglas fir megagametophyte developm ent... Endogenous levels o fZ and ZR, on a fresh weight basis, during Douglas fir megagametophyte developm ent... Endogenous levels of iP and iPA. on a drv weight basis.
107
109
110
12 during Douglas fir megagametophyte developm ent...
Endogenous levels o f iP and iPA, on a fresh w eight basis, during Douglas fir megagametophyte developm ent 113
basis, during Douglas fir megagametophyte development 114 F ig u re 21 Endogenous levels o f lAA and lAAsp, on a fresh weight
basis, during Douglas rlr megagametophyte development 116 F ig u re 22 Endngennus levels o f ABA and .AB.A-GE, on a d r/ weight
basis, during Douglas fir m egagam eiophue development 117 F ig u re 23 Endogenous levels o f .ABA and .AB.A-GE. on a fresh weight
Completing this study in the Biology Department at the University o f Victoria is one o f the most fulfilling experiences o f my life, and this would have been im possible without the help o f the following people. I will forever he gratefi!! to my super'.'isor. Dr. Patrick von .Aderkas. for his guidance, unwavering support and consideration throughout my studies as well as in editing this dissertation. 1 w ish to thank Drs. Louise Page. Barbara Hawkins and fe rn ' Pearson, my committee members, for their constm ctive suggestions throughout my study and on this dissertation.
1 am deeply grateful to Dr. Joe W ebber o f the BC Ministr}' o f Forests at the Glyn Road Research Station for providing the plant material for this study and Dr. Philippe Label for supplying the conjugated hormones and antibodies and their working dilutions for the hormonal analysis experim ent. I wish to thank Drs. N'anc} Sherwood and Will Hintz for allowing me to use their laboratory equipment for some o f my experim ents. Heather Down and Tom Gore for their assistance in the .Advanced Imaging Laboratory. Luke Chandler for taking the time out o f his studies to help me with the plates in this dissertation and M arlies Rise for her help with the em bedding and sectioning o f specimens (You are a true angel Mar.!!).
I wish to thank Cathy Leaiy, who was a work study student in our laboratory, for her help with the m egagam etophyte culture study. I also w ish to thank B rett Poulis, Dan O 'n eil and Carol for their help with the Beckman HPLC, my lab mate Stephen O 'L eary for ju s t being Stephen, and the people in the Graduate Centre for Forest Biology for
department. Finally, I wish to thank my friends Brian . Nadja. Erica Fradinger, Marlies Rise, my family m em bers Doug Chiwocha. Cleopatra Makoni and little Tim m \' for always being there tor me. The financial support for this study was provided by the operating grant o f rhe Natural S c i e n c e and E n g i n e e r i n g Research Council o f Canada held bv Dr. Patrick von .Aderkas.
INTRODUCTION
Douglas fir {Pseiidoisuga menziesiî) is a m em ber o f the Pinaceae and has a wide distribution in the Pacific Northwest. Its natural range extends from the southern h alf o f British C olum bia through the United States into M exico (Owens, 1973). Because o f its economic im portance as a lumber species, methods to ensure a constant supply o f superior reforestation stock which can be produced in a short period o f time are necessary.
The reforestation o f many conifer species has encountered several impediments. Conifers have long juvenile periods (10-20 years), before they become reproductive (Pharis and King, 1985). Gibberellic acid (GA) can dram atically shorten the juvenile period such that flowering can be induced in plants 3-12 months old (Pharis and King,
1985). D ifferent classes o f GA are required to induce flowering in different conifer families. Polar G As such as G A3 are useful in species belonging to the Taxodiaceae and Cupressaceae whereas non-polar G As such as GAg and GA4 / 7 are effective in m embers of the Pinaceae (Pharis and King, 1985), including Douglas fir (Owens, 1991). This effect o f G As has been widely exploited in com mercial forestry w here it is used to obtain seed from young trees that have not developed to the reproductive phase.
A nother common problem is the periodicity o f cone crop production, w hich in Douglas fir has been found to average five years (Owens, 1973). GA used alone or in conjunction w ith other treatments that promote flowering can overcom e this obstacle in seed orchards. Girdling, which is believed to lim it the dow nw ard transport o f
carbohydrates from the crown, enhances flowering in several species (see Owens, 1991 for review). M oisture stress treatment, which is achieved by either drought or root
for review).
An additional problem is that Douglas fir has a reproductive cycle that spans seventeen months, lim iting the harvesting o f mature seed to once every two years (Owens, 1973). Production o f seedlings from other methodologies, such as in vitro culture, may increase the production o f selected genotypes in a shorter period, and in turn, speed up the reforestation process for Douglas fir. These novel breeding strategies include m icropropagation, somatic embryogenesis and in vitro fertilization (IV F). One form of m icropropagation which has been successful in both conifers and angiosperm s is organogenesis, in w hich foliage explants, e.g. individual conifer needles, are initially cultured to stim ulate adventitious shoot development. The shoots are then rooted, allowing plants to be produced in vitro (see Bonga and von Aderkas, 1992 for review ). Somatic em bryogenesis is a form o f vegetative propagation in which em bryos are produced from cells that are not naturally embryogénie following induction o f the primary explants (usually im mature zygotic em bryos in conifers) with plant horm ones (see Bonga and von A derkas, 1992 for review).
In vitro fertilization requires the introduction o f a male gamete to a fem ale gamete and is therefore a form o f sexual propagation. The type o f IV F in conifers Involves direct microgametophyte / m egagametophyte interaction. Steps tow ards in vitro production of embryos from IVF were made recently in Douglas fir (Takaso et al, 1996; D um ont- Béboux and von Aderkas, 1997; Fem ando et al, 1997 ). The first report o f successful IVF in any conifer was by Fem ando et al. ( 1998) in Douglas fir. Although this is a m ajor breakthrough in conifer research, the ultimate goal is to rescue the em bryos resulting from this technology and grow them into plants. Em bryo rescue is a m ethod that is used to raise em bryos that would otherwise abort due to post-zygotic selection (R am m ing,
1990). In conifers, em bryo rescue o f zygotic and proem bryonic stages has n ot been accomplished (Sterling, 1949; Gates and Greenwood, 1991). In part, this co uld be due to
levels o f plant hormones.
Plant hormones are defined as “a group o f naturally occurring organic com pounds w hich influence physiological processes at low concentration” (D avies, 1995). Nine classes o f organic substances have been identified that fall under this broad definition. They are auxins, gibberellins, cytokinins, ethylene, abscisic acid, polyam ines,
jasm onates, salicylic acid and brassinosteroids. For historical reasons, most studies have concentrated on the first five hormones. These have been shown to affect physiological processes that occur during seed developm ent in several species o f angiosperm s and conifers. O f these five horm ones, abscisic acid (ABA) is the m ost studied. Very few studies have investigated changes in plant hormone levels during seed developm ent in conifers. ABA levels have been quantified during zygotic em bryogenesis in loblloly pine (K apik et al., 1995) and more recently in white spruce (Carrier et al., 1999). Endogenous auxin, gibberellins and ABA levels in developing white spruce seeds were published by Kong et al. (1997).
Embryo rescue o f early stage embryos requires the optim ization o f culture m edia and culture conditions to prom ote further growth o f the em bryos. Prior know ledge o f endogenous hormone levels may be invaluable in the design o f culture m edia for use in such studies. Imitation o f conditions in the natural seed environm ent may prom ote further developm ent o f the cultured em bryos (Ramming, 1990). Techniques w hich are currently available for the quantification o f plant hormones include high pressure liquid
chrom atography (HPLC), gas chrom atography (GC), gas chrom atography-m ass
spectroscopy (GC-M S) and enzym e immunoassays (Els) (H ock et al.. 1992). Because of the small amounts o f horm ones in plant tissue samples, methods w ith high sensitivity are required for quantification. The high affinity and specificity o f Els m akes them
restricted to investigating the role played by ABA in the establishm ent o f dormancy in mature seeds (Jarvis et al., 1997; Bianco et al., 1997). Hormone levels during earlier stages have not been studied. This inform ation is essential in light o f the recent report by Fem ando et al. (1998) o f IV F in this species. The reported success o f somatic
em bryogenesis technology in conifers (see Bonga et al., 1995; B ecw ar and Pullman, 1995; C handler and Young, 1995 for reviews) could be applied to em bryos resulting from IVF studies. The rescued em bryo could be multiplied by this method, providing a quick way for propagation.
M y research focused on megagametophyte developm ent and was designed to address the following questions:
Question 1. Do plant hormone levels change during m egagametophyte development? Question 2. Are m orphological and physiological changes during developm ent
independent o f one another i.e. are there m orphological markers of physiological change?
Question 3. Can m egagam etophytes be cultured prior to fertilization ?
Question 4. Can em bryos be rescued from megagametophytes cultured soon after fertilization?
LITERATURE REVIEW
2.1 Conifer seed development
Em bryogeny starts with the establishment o f the haploid m egagam etophyte well before fertilization. This is follow ed by the formation o f structures bearing the female gametes (the eggs) within the megagametophyte tissue (Figure la). In conifers, fertilization is a single fusion event between a male nucleus and the egg nucleus. The embryo initially develops within the archegonium (Figure lb) until suspensor elongation pushes it into the m egagam etophyte (Figure Ic) w here it matures surrounded by the haploid prothallial cells (Figure Id; Singh. 1978).
2.1.1 Embryogeny
The developm ental events leading to egg cell formation in the prefertilization megagametophyte have been extensively studied by light and electron m icroscopy for various conifers (Allen and O w ens, 1972; Owens and M orris, 1990; Runions and Owens,
1999a; see Singh, 1978 for review). Briefly, developm ent starts with differentiation o f the megaspore m other cell in each ovule. This cell undergoes meiosis, resulting in four daughter cells, three o f w hich degenerate. The rem aining cell, the functional m egaspore, undergoes a num ber o f free nuclear divisions, giving rise to a large m ultinucleate cell. The nuclei are suspended in a thin layer o f parietal cytoplasm surrounding a large central vacuole. Cellularization o f the ceonocyte follows in a process called alveolation (Singh,
1978). Archegonial initials differentiate at the m icropylar end o f this structure. These cells enlarge and divide to form a small primary neck cell and a larger central cell. The primary neck cell undergoes a few mitotic divisions to form a short neck. The central cell
(p) is trapped in the micropyle (m) where it elongates while m egagametophyte developm ent progresses. The haploid m egagam etophyte is surrounded by the megaspore w all (mw) and is made up o f prothallial cells (pc) and the archegonia (a) located at the micropylar end. The archegonia contain central cells prior to fertilization and mature egg cells at fertilization. The wall o f the ovule, the seed coat (sc) develops from the integument. The nucellus (n) is a diploid tissue through w hich pollen tubes grow in order to insem inate the eggs at fertilization.
Figure IB A fter pollen germination, the pollen tube (pt) grows through the nucellus (n) to penetrate the archegonium, releasing the male gametes at fertilization. The zygote divides to give rise to the proembryo (pe) within the archegonium.
Figure 1C The nucellus begins to degenerate as elongation o f the suspensor (s) pushes the em bryos (e) out o f the archegonium and into the prothallial cells o f the m egagam etophyte before corrosion cavity formation.
Figure ID Cells in the central region o f the m egagam etophyte degenerate to form the
corrosion cavity (cav) as the suspensor pushes the dom inant em bryo deeper into the m egagam etophyte tissue. The embryo m atures in this cavity, surrounded by the haploid prothallial cells.
PC
sc
C
av
both housed w ithin an archegonium (Allen and Owens, 1972; see Singh, 1978 for review). The prothallial cells surrounding the archegonia differentiate to form the jacket cells. The num ber o f archegonia formed is species dependent w ithin a conifer family, ranging from 1 up to 100 in some species (see Singh, 1978 for review). In Pinaceae for example, at fertilization, each Douglas fir megagametophyte has 4-6 archegonia (Allen and Owens, 1972), Larix has 3-4 (Schopf, 1943) whereas interior spruce has 2-3 (Runions and O w ens, 1999a). Podocarpus totara, which is a m em ber o f the
Podocarpaceae family, has 4-6 archegonia ( W ilson and Owens, 1999) and in Taxaceae,
Tc l x u s brevifolia has an average o f 4 archegonia per ovule (Anderson and Owens, 1999). The m egagam etophyte may be involved in more than ju st egg production. Regulation o f pollen behavior has been implied in a series o f recent studies. In Douglas fir, the presence o f an ovular secretion in the micropyle prior to fertilization has been reported by T akaso et al. ( 1996) and von Aderkas and Leary ( 1999). Formation o f the ovular secretion was dependent on the developmental stage o f the megagametophyte. Takaso et al. ( 1996) found that the secretion was absent in ovules containing central cell stage m egagam etophytes, only being produced a week before fertilization when the egg cells were nearly mature. However, von Aderkas and Leary (1999) recently reported the appearance o f this secretion for a two week period, starting as early as the central cell stage. In this species, pollination occurs about two months before fertilization (Owens and Morris, 1990), when megagametophyte developm ent is ju st beginning. The pollen grains rem ain trapped in the micropyle while megagametophytes progress through the various developm ental stages leading to egg cell formation. During this time, the pollen grains shed their exines and slow ly elongate over a period o f 6 weeks (Owens and Morris, 1991). A n hypothesized secretion from the egg and prothallial cells into the micropyle cau sed dissolution o f the pollen grain wall concurrent with pollen tube formation (Takaso and Owens, 1994; 1996; Takaso et al., 1996). Owens and M orris
pollen tube initiation. Cessation o f fluid production ju st before fertilization (Takaso et al., 1996) was accom panied by the appearance o f transparent areas throughout the egg
cytoplasm , suggesting that the secretion contained material resulting from cytoplasm ic degradation o f the egg cells (Takaso and Owens, 1994). Indirect evidence that the secretion originated from the m egagam etophyte was provided by Takaso et al. (1996) in their experim ent with m egagam etophyte homogenates. Sim ilar ovular secretions have been reported in LarLx (B am er and Christiansen, I960; Said et al., 1991; Owens et al.
1994). Although conifers are generally not believed to have any prezygotic selection m echanism s (Owens, 1991), Takaso and Owens, ( 1994) reported a high percentage of distorted pollen grains in self-pollinated ovules com pared to cross-pollinated ovules, suggesting that there may be some form o f pollen recognition m echanism prior to pollen tube germination.
Events occurring in the archegonia o f several conifers ju st before as well as soon after fertilization have been described ultrastructurally (Owens and M orris, 1990; 1991; Runions and Owens, 1999a; 1999b; Anderson and Owens. 1999) and by light microscopy (A llen and Owens, 1972, Wilson and Owens, 1999). In particular, cytoplasmic
inheritance of chloroplasts and m itochondria in Douglas fir has been studied (Owens and M orris, 1990; 1991). As a m echanism, these authors suggested that transform ation o f plastids into large inclusions, which begins in the central cell and is com plete at the m ature egg cell stage, functions to exclude the m igration o f m aternal plastids into the early em bryo. In interior spruce, this is followed by further spatial isolation o f the
m odified plastids from the perinuclear zone that surrounds the egg nucleus (Runions and O w ens, 1999a). M aternal plastid transform ation prior to fertilization has also been reported loxPinus and L arcr (Cam efort, 1962; 1967 respectively, cited in Owens and M orris, 1990) as well as Taxus (A nderson and Owens, 1999). Thus chloroplasts are exclusively o f paternal origin in the above genera. A lthough aggregation o f m itochondria
in the perinuclear zone o f the egg nucleus prior to fertilization results in a large maternal contribution o f this organelle into the em bryo, approxim ately 10% o f m itochondria may be o f paternal origin in Pinus (W agner et al.. 1992) and Pseudotsuga (O w ens and Morris,
1991). M ixed inheritance o f m itochondria was also recently reported in Taxus brevifolia by Anderson and Owens ( 1999). In Picea, genetic studies have confirm ed that
mitochondria are strictly o f m aternal origin (Sutton et al., 1991; David and Keathley, 1996). This result agrees with what was suggested from an ultrastructural study by Runions and Owens ( 1999b).
Once formed, the pollen tube must penetrate the nucellus before it can enter the egg. The m echanism by w hich a pollen tube grows through the nucellus to find the neck of an archegonium has been the subject o f much speculation. Runions and Owens (1999a) reported lipid accum ulation in the archegonial cham ber above each neck which may direct pollen tube grow th tow ard an archegonium, although they did not think that such a signal was strong enough to transverse the nucellus. Indeed, neck cells have characteristics o f secretory cells in D ouglas fir (Owens and Morris, 1990). However, results from a recent study by Dum ont-Beboux et al. ( 1998) did not support this proposed function o f the neck cells. The pollen tube penetrated the egg cell through the side o f the archegonium in vitro (D um ont-Beboux et al.. 1998). Furthermore, A nderson and Owens (1999) reported degeneration o f neck cells prior to fertilization in Taxus brevifolia. In angiosperms, directional grow th o f pollen tubes toward the ovary is believed to be a chemotropic response (for review, see M ascarenhas, 1993). W hether a chem ical signal does exist in conifers and w hether it is produced by the neck or egg cells o r both has yet to be determ ined. The pollen tube tip releases many proteins with hydrolytic properties, which aid pollen tube penetration o f the nucellus (Pettitt, 1985; Runions and Owens.
1999a). Shafer and K riebel (1974) reported that nucellar cells around the grow ing pollen tube showed an increase in cytoplasm ic RNA and suggested that the pollen tubes may
produce a horm one-like substance w hich stim ulates RNA and protein synthesis in these ceils.
2.1.2 Fertilization and zygotic embryogenesis
The fertilization process in angiosperm s is different from that in conifers. In angiosperm s, seed developm ent is a result o f double fertilization (Taiz and Zeiger, 1991). One male gam ete fuses with an egg cell, w hich is housed within an em bryo sac, to form the zygote. The second unites with the diploid central cell of the em bryo sac to form a triploid endosperm, w hich develops into a nutritive tissue. In conifers, only one pollen tube will penetrate each egg cell, and the deposition o f a pectinaceous substance at the site o f pollen tube entry into the egg may inhibit insemination by other pollen or the release o f egg contents (Runions and O w ens, 1999b). Ultrastructural studies o f
fertilization in conifers have dem onstrated that although two male gam etes are deposited into the egg cell upon pollen tube penetration, only the leading nucleus fuses with the egg nucleus, while the trailing gamete remains in the m icropylar end o f the egg cell
cytoplasm and eventually degenerates (Owens and M orris, 1990: Runions and Owens. 1999b: W ilson and O wens, 1999: see Singh. 1978 for review). Pollen tube penetration is prim arily through the neck cell of each archegonium , although Runions and Owens ( 1999b) reported one instance when a pollen tube grew through the side o f an egg cell.
In the Pinaceae, mitotic divisions o f the zygote nucleus gives rise to four free nuclei which m igrate to the chalazal end o f the archegonium and are arranged in a single tier. Divisions by these nuclei quickly give rise to eight nuclei. Cell wall formation results in a two tiered proem bryo, in w hich the upper cells are open to the egg cytoplasm. In Douglas fir, divisions by cells in the low er tier results in a mature proembryo
com prised o f 12 cells in three tiers: an open tier, a suspensor tier in the middle and a low er embryo tier (A llen and Owens, 1972: Owens and Morris. 1991). In most Pinaceae, the matiu’e proem bryo is com prised o f 16 cells in 4 tiers (see Singh, 1978 for review).
Elongation o f the suspensor cells pushes the early em bryo into a corrosion cavity which forms in the megagametophyte tissue soon after fertilization. The corrosion cavity is absent in Douglas fir megagametophytes before suspensor elongation (O wens et al.,
1993). Although the m echanism o f cavity formation is still unknown, Shafer and Kriebel (1974) described an increase in cytoplasmic RNA accum ulation by prothallial cells in the region o f the megagametophyte that later formed this fluid filled structure in eastern white pine. This increase was evident in pollinated megagametophytes at o r before fertilization, as well as in pollinated unfertilized explants (in which the egg cells were degenerating) collected several days past fertilization period. The conifer em bryo develops further and matures inside the corrosion cavity (Singh, 1978).
Detailed descriptions o f zygotic em bryogenesis in conifers have been published (see Singh, 1978 for review). This process in Douglas fir has been thoroughly described at the light microscope level (Allen, 1946: 1947a; 1947b, Allen and O wens, 1972, Grob et al., 1999) and at the ultrastructural level (Owens et al., 1993). Developm ent can be divided into the proembryo, early, late and mature em bryo stages (Allen and Owens,
1972). The term proembryo encompasses the free-nuclear and cellular em bryo stages within the archegonia prior to suspensor elongation. This is followed by the early embryo stage, in w hich embryonal masses are pushed into the corrosion cavity following
suspensor elongation (Krasowski and Owens, 1993). Further divisions result in a club shaped embryo, in which polarity is established subsequent to developm ent o f the
proxim al and distal meristems (Krasowski and O wens, 1993). H istodifferentiation occurs during the late embryo stage. This period is characterized by formation o f the root and shoot apical meristems, developm ent o f the hypocotyl-shoot axis and by the initiation and elongation o f cotyledons (Krasowski and Owens, 1993). Further developm ent results in the m ature dormant em bryo (Allen and Owens, 1972).
2.1.3 Storage reserves during conifer seed development
Prior to fertilization, conifer megagametophytes contain little storage m aterial and are translucent. A ccum ulation of storage products com mences during early
embryogenesis (Schopf, 1943). Although the major storage reserves in mature conifer seeds are lipids (O w ens et al., 1993; Stone and Gifford, 1999), a majority o f studies have concentrated on biochem ically characterizing the nature o f storage proteins in m ature seeds (Lam m er and G ifford, 1989; Hakman et al., 1990; Baker et al., 1996; M isra and Green, 1994) and m obilization of these proteins during germ ination (Gifford et al., 1989; Lammer and Gifford, 1989; Gifford and Tolley, 1989; King and Gifford, 1997). It has been suggested that studies focusing on seed storage proteins are invaluable because they provide insight into D N A regulated events during em bryogenesis (Owens, 1995).
H istochem ical and ultrastructural studies focusing on the accumulation o f storage products in conifer m egagam etophytes and em bryos during em bryogenesis have been published for Douglas fir (Owens et al., 1993) and white spruce (Krasowski and O w ens,
1993). Starch was the only storage product detected in the corrosion cavity region o f megagametophytes at fertilization through proembryo developm ent in Douglas fir (Owens et al., 1993). The archegonia may initially provide nourishm ent for the developing proem bryo and early embryo through the suspensor (Owens and M orris,
1991). Once the early em bryo is forced into the prothallial tissue by suspensor
elongation, rapid hydrolysis o f starch in the corrosion cavity region occurred (O w ens et al., 1993). Lipid and protein deposition was first detected in megagametophytes
containing early em bryos. By the club-shaped embryo stage, protein and lipid content had increased in the com bined em bryo and megagametophyte. A fter histodifferentiation, when cotyledons and the root cap had form ed and the m eristem s w ere established, starch accumulation was prim arily in the root cap w hile lipids and proteins were more
Similar accum ulation patterns were reported for Picea glauca (Krazowski and Owens, 1993).
M ost m olecular and biochemical studies com m ence after histodifferentiation (Owens et al, 1993) and these studies characterize the nature o f storage proteins in conifer seed. G ifford ( 1988) identified crystalloid proteins in mature seeds o f Pinus monticola and several other Pinus species as the major storage proteins. The wide
occurrence o f crystalloids in mature conifer seeds has since been dem onstrated by similar studies in lodgepole pine (Lam m er and Gifford. 1989); N orw ay spruce (Hakm an et al.,
1990), interior spruce (Flinn et al., 1991), Douglas fir (M isra and Green, 1990: 1991) and eastern white pine (Baker et al., 1996). A large proportion o f these proteins is located in the m egagam etophyte and the remainder in the embryo.
M isra and G reen ( 1991) showed that crystalloid protein synthesis is developmentally regulated during embryogenesis in Douglas fir, with m aximum synthesis occurring during the mid-to-late cotyledonary stages. Similar results were obtained for interior spruce by Flinn et al. ( 1991). N orthern blot analysis with
megagametophytes at various stages of development found mRNAs encoding crystalloids were maximal from the club-shaped through to cotyledonary stages and then declined during later m aturation in both the embryo and the m egagam etophyte (Leal and M isra,
1993). Sim ilar results were obtained by Baker et al. ( 1996) in eastern white pine.
It appears that crystalloid gene expression is both developm entally regulated and tissue specific. In Douglas fir, mRNAs encoding crystalloids were not detected in mature dry seed o r in leaves and roots o f 2 week old seedlings by D NA-RNA hybridization ( Leal and M isra, 1993). Baker et al. (1996) also found differential gene expression within the ovules; mRNAs encoding crystalloids were not detected by either D N A -RN A
hybridization, tissue printing or RNA filter hybridization in sporophytic tissue (i.e. inner and outer integum ents as well as the nucellus).
The evolutionary relationship between crystalloid proteins o f gym nosperm s and the 1 Is globulins o f angiosperms has been investigated. W hite spruce crystalloids were im m unologically related to those o f several conifers in Pinaceae and to I Is globulins o f several angiosperm s (M isra and Green, 1994). Baker et al. ( 1996) also found mRNAs in eastern white pine ovules that had significant sequence homology to the m RN A s o f 1 Is globulins o f several angiosperms. There was no immunological relationship betw een 1 Is globulins o f legumes (legumins) with crystalloid proteins o f white spruce (M isra and G reen, 1994) o r Pinus (G ifford, 1988). However, a com parison o f legumin precursor cD N A w ith Douglas fir crystalloid cDNA showed 29-38.5% identity, suggesting that these storage protein genes may have a com m on ancestry but have diverged during the course o f evolution (Leal and M isra, 1993).
In addition to storage proteins, deposition o f sugars and lipids also occur during conifer seed development. Gates and Greenw ood (1991) reported that increases in hexose sugar, soluble protein and lipid concentration paralleled the increase in seed dry w eight from the club-shaped em bryo to the early cotyledonary stages during em bryogenesis o f Pinus resinosa. An ultrastructural and histochemical study by Krasowski and Owens ( 1993) yielded sim ilar results for protein and lipid accum ulation during seed developm ent in white spruce. However, a more recent study in white spruce utilizing gas-
chrom atography detected lipids prior to fertilization, when m egagam etophytes were still aqueous (C arrier et al.. 1999). Stone and G ifford ( 1999) recently reported that lipids com prised 59% o f the total storage reserves in mature seeds o f loblolly pine.
2.2
In vitro
megagametophyte culture studiesThe conifer megagametophyte is com prised of a variety o f cells, nam ely the prothallial, jacket, neck and either central cells prior to fertilization or egg cells at fertilization. Few studies have been published that were carried out to establish culture conditions for IV F o f megagametophytes isolated prior to fertilization. D ouglas fir
m egagam etophytes cultured prior to fertilization on m edia supplem ented w ith auxin and cytokinin grew to lengths not normally achieved in situ by 9 days in culture (von Aderkas et al, 1997). M a et al. ( 1998) reported m ultiplication o f prothallial, neck and jacket cells in cultured Douglas fir megagametophytes isolated at fertilization. However, egg cell viability could only be maintained for 3-4 days in the same species (Fernando et al.,
1997). Fernando et al. ( 1997) and von Aderkas et al. ( 1997) also found that grow th of prothallial cells continued even after the egg cells had degenerated.
G ates and Greenwood (1991) reported that fertilized m egagam etophytes o f Pinus resinosa cultured on media supplem ented w ith sucrose concentrations o f up to 21 % resulted in m egagam etophytes that resembled material in situ. H owever, they found that developm ent o f em bryos within the m egagam etophytes was arrested on these same media.
Until recently, the primary focus o f in vitro culture o f m egagam etophytes in conifers has been the induction o f haploid cultures for em bryogenesis. This is historical, partly because o f the successful creation o f haploid based breeding program s for
angiosperm s. Haploid plants have been produced from both anthers and unfertilized ovules in angiosperm tree species, (see Chalupa. 1995: Gosal et al.. 1995 for reviews). In gym nosperm s, the first and only tissue to produce haploid cultures was
m egagam etophytes (Bonga et al.. 1994; see Bonga and von A derkas, 1992 for review). A lthough seeds o f selected genotypes are currently produced in seed orchards from tree im provem ent programs, the production o f breeding lines that are hom ozygous for desirable traits by inbreeding is restricted by the long juvenile periods and the long reproductive cycles in conifers. In vitro production o f haploid plants would therefore be a particularly valuable system in conifers (B onga and Fowler, 1970; Sim ola and Honkanen,
1983; Pattanavibool et al., 1995). It has been suggested that the haploid plants could be diploidized artificially by colchicine treatm ent, allow ing the production o f homozygous breeding lines (B onga and Fowler, 1970). A flow cytom etric and chrom osom e study by
Pattanavibool et ai. (1995) reported diploidization after 2-9 years in culture o f haploid lines o f L. decidua m egagam etophytes. These authors em phasized the benefits o f regenerating plants from these diploidized cultures as this w ould eliminate the need to artificially double chrom osom e num ber as originally suggested by Bonga and Fow ler (1970).
Haploid callus tissue proliferation has been induced in a number o f conifer species from explants o f megagametophytes grown on m edia supplem ented w ith exogenous plant horm ones. M ajor factors that had an im pact on successful induction of em bryogénie cultures from such explants were genotype and collection date w ithin a species. Bonga and Fow ler ( 1970) obtained haploid callus tissue from m egagam etophyte explants o f Pinus resinosa isolated between fertilization and em bryo elongation only. In L decidua, em bryogénie callus was produced with m egagam etophytes isolated a few w eeks after fertilization (von Aderkas et al., 1987).
In Picea abies, culture o f megagametophytes from im m ature seed resulted in callus tissue with a poo r capacity for organogenesis (Sim ola and Honkanen, 1983). In a later study using the sam e species, Hakman et al. (1985) were unable to induce haploid callus formation from m egagam etophyte explants. Further studies in P. abies revealed the importance o f horm one levels and nitrogen source on haploid em bryoid regeneration from m egagam etophyte callus although the embryos they produced failed to m ature in vitro (Sim ola and Santanen, 1990). Successful production o f haploid em bryoids which were able to develop into em bryos was first achieved in iM rix decidua (N agm ani and Bonga, 1985). W ork in other species o f Larix also produced haploid plantlets b ut these only exhibited lim ited grow th after potting (von Aderkas et al., 1990). Further studies in L decidua reported the first successful production of plants from a haploid
m egagam etophyte-derived culture line that later survived greenhouse conditions (von A derkas and Bonga, 1993).
2.3 Embryo rescue
In plants, post-zygotic selection can cause poor em bryo developm ent (Ram ming, 1990). This is evident in som e interspecific crosses in w hich the hybrids fail due to em bryo abortion. Post-zygotic selection has been overcom e in fruit breeding program s by the use o f embryo culture (see Ramming, 1990 for review). Seeds fail to develop due to cessation o f embryo developm ent soon after fertilization in seedless varieties o f grapes (Em ershad and Ramming, 1984). In early ripening Primus (Ramming, 1985) spontaneous abortion results in the production o f seedless fruit. Breeding plants from such genotypes have been produced by culturing embryos prior to abortion. Because o f the small size o f the em bryos, whole ovules were initially cultured. This allowed further developm ent o f the em bryos within the ovule which were then subcultured and grown into plants (Ram m ing, 1990). D epending on the genotype, multiple embryos developed in some cultured ovules as a result o f somatic em bryogenesis by the zygotic em bryos (Ram m ing,
1990). Such work has been successful in early ripening Primus (R am m ing, 1985) and seedless varieties o f Vitis (for review, see Ramming, 1990).
A few recent reports on work done in maize have opened up the possibility o f using this technology to rescue embryos resulting from in vitro fertilization (IVF) experim ents. The first report o f IVF in maize was by Kranz et al. ( 199 la . b). Since then, em bryo regeneration into plants has been achieved by in vitro culture o f em bryo sacs soon after fertilization in this angiosperm (Cam pem ot et al., 1992; Mol e t al.. 1993; 1995; M atthys-Rochon et al., 1998). This latest study by M atthys-Rochon et al. ( 1998)
validated the significance o f the types and the levels o f hormones used, the genotype, as well as the carbon source on successful plant regeneration.
In conifers, seedlings were produced by culturing mature em bryos resulting from interspecific crosses betw een the blister rust susceptible Pinus lambertiana w ith pollen from the blister rust resistant P. annandi or P. koraiensis (Stone and D uffield, 1950). H ealthy seedlings were also reported from cultured em bryos o f P. lambertiana (Sacher,
1956; H addock, 1954). In larch, Sterling (1949) attempted em bryo rescue by isolating early em bryos from m egagametophytes and only obtained limited growth after two months o f culture. He also observed cleavage em bryogenesis by a few o f the cultured larch em bryos, w hich does not occur in situ. This result led Sterling (1949) to suggest that rem oval o f the megagametophyte m ay have released the em bryos from
developm ental constraints that em bryos in situ would naturally encounter during
developm ent. Based on work in Pinus resinosa, osm otic potential in the corrosion cavity o f the m egagametophyte does not appear to play a critical regulatory role during em bryo developm ent. Gates and Greenwood, ( 1991) reported that only slight changes occurred in supernatants obtained from megagametophyte containing em bryos at the zygote up to cotyledonary stages. Although em bryo rescue o f early stages has not been achieved in conifers, trem endous progress has been made in developing somatic em bryogenesis system s in a number o f economically im portant species. Immature zygotic em bryos are com m only used as the primary explants for conifers making these studies a form o f em bryo rescue.
2.3.1 Somatic embryogenesis
Somatic embryogenesis in conifers was first reported by Hakman et al. ( 1985) in Picea abies from cultures initiated from immature zygotic embryos. Since then, the focus has shifted to optimizing culture conditions for a number o f different explant types for several conifer species. Toward this goal, a num ber o f researchers have attem pted to initiate somatic em bryogenesis from mature seeds as well as from cotyledons obtained from germ inated em bryos (See Becw ar, 1993 for review). This technology is particularly attractive for conifers because it provides a way to shorten the long reproductive cycles, which range from 2 years in some species e.g. Douglas fir to 3 years in Pinus. The potential to com m ercialize somatic em bryogenesis by using it together w ith tree im provem ent programs in conifers provides a means for mass propagating em bryos
resulting from controlled crosses o f genetically superior parents (Adams et al., 1994; see Becwar, 1993 for review).
Somatic em bryogenesis can be divided into four steps; I) initiation 2) proliferation 3) m aturation 4) germ ination (Becwar, 1993; von A rnold et al., 1996). Initiation o f som atic embryos from cells that are not naturally em bryogénie requires induction with p lant hormones. The transition is m ediated by auxins and cytokinins (Dodeman et al., 1997). Culture o f im m ature zygotic em bryos on m edia supplem ented with auxins and cytokinin produced em bryogénie tissue in Picea abies (Hakman et al.,
1985; Simola and Santanen, 1990), larch (von Aderkas et al., 1990; Thom pson and von Aderkas, 1992), Picea glauca (D unstan et al., 1988), Pinus taeda (B ecw ar et al., 1990) and in Douglas fir and Pinus taeda (G upta et al., 1988). It appears that auxin and
cytokinin mediate em bryogénie tissue initiation by affecting cell polarity and prom oting asymmetric cell divisions in the prim ary explant (reviewed in Dodeman et al., 1997). Repeated cell divisions by some cells in the primary explant results in the formation o f dense clusters o f cells which give rise to somatic em bryos (von Arnold et al.. 1996). Although a m ajority o f studies utilize im mature zygotic em bryos as the primary explant. Jalonen and von A rnold ( 1991) reported somatic em bryo formation from mature em bryos o f P. abies.
The initiated somatic em bryos continue to divide and give rise to em bryogénie cell lines in the proliferation step (von A rnold et al., 1996). During maturation,
proliferation stops and the embryos grow in size and accum ulate storage products (von A rnold et al., 1996). Fliim et al. ( 1991) found that som atic and zygotic embryos
accumulated the sam e storage proteins during developm ent in interior spruce. M aturation requires removal o f somatic em bryos from auxin and cytokinin containing media and placem ent on m edia supplemented w ith A B A (Jalonen and von Arnold. 1991; Lelu et al.,
1995; Dunstan et al., 1988; Attree et al., 1992; Becw ar and Feixer, 1989; Becw ar et al., 1990; Gutman et al., 1996). Attree et al. ( 1992) reported that high osm oticum together
w ith A BA resulted in em bryos that closely resembled mature zygotic em bryos in lipid content for P. glauca. Lipid accum ulation was also increased in loblolly pine somatic em bryos cultured on m edia containing A BA (Becw ar and Feirer, 1989). Culture o f som atic em bryos on ABA w ithout osm oticum resulted in precocious germ ination during late m aturation, and this was attributed to a decrease in sensitivity to ABA (Attree et al.,
1992). Desiccation o f the mature em bryos improved plantlet recovery in hybrid larch (Lelu et al.. 1995) as well as in white spruce (Attree et al.. 1992). The first repon of successful plant regeneration from haploid em bryogénie cultures in conifers was made by von Aderkas and Bonga ( 1993) in L. decidua, which lends credence to the use o f this system in alternative breeding programs for conifers.
2.4 Plant hormones during seed development
2.4.1 Background
Studies investigating horm one action in plants have been limited to either the effect o f exogenous hormones or quantification o f endogenous levels followed by attem pts to correlate this with physiological processes. W hile these studies have
contributed invaluable inform ation on the effects o f different classes o f plant hormones, the signal transduction pathways have remained a mystery. Signal transm ission to the nucleus may be through any one o f several systems, including G TP binding proteins, protein kinase cascades and m em brane ion channels. H orm ones may change the activity o f D N A -binding proteins w hich in turn may change the expression o f genes directing developm ental program s (M ulligan et al., 1997). The recent advent o f m olecular genetic studies w ith Arabidopsis mutants has resulted in the identification o f genes involved in the signal transduction pathways o f ethylene (see Lelièvre et al., 1997 for review) and A B A (C utler et al., 1996; Pei et al., 1998; see M cCourt, 1999; Leung and G iraudat, 1998 for reviews). A lthough these pathw ays have not been fully described, researchers now
have a new tool for further elucidation o f plant horm one signaling pathways. In addition, the m ajor focus has been on angiosperm species and as a result, there is a dearth o f inform ation in the literature on the role o f plant horm ones during conifer seed developm ent. The studies in angiosperms are reviewed in the section below.
2.4.2 Angiosperm seed development
Em bryo developm ent in an angiosperm starts w ith an asymmetric transverse division by the zygote which establishes polarity soon after fertilization (M ayer et al.,
1993). The em bryo develops from the distal cell and the suspensor from the proxim al cell. Fertilization is followed by a period o f rapid growth and differentiation by the em bryo (Rock and Quatrano, 1995). Em bryogenesis can be divided into four phases; the proem bryo, globular, heart-shaped and finally the torpedo stages. Prior to suspensor formation, the em bryo is in the proem bryonic stage. Further divisions result in the formation o f the globular stage em bryo with an attached suspensor. The heart-shaped stage occurs following divisions by the m eristem atic cells o f the cotyledons. Torpedo- stage em bryos results from elongation o f the cotyledons as the apical and root tips are
form ed (Raven et al., 1986).
Em bryo developm ent is under the influence o f the surrounding seed and m aternal environm ents as well as under genetic control (Xu and Bewley, 1991). Some o f the
various gene sets that are expressed at different stages during embryogenesis and seed developm ent have been identified (M ayer et al., 1993; M einke et al., 1994; reviewed in G oldberg et al., 1989; reviewed in Dodeman e t al.. 1997). In maize, 51 em bryo-specific m utations o f genes crucial to the morphogenesis o f the em bryo at various developm ental stages have also been isolated and characterized (Clark and Sheridan, 1991). During the m aturation phase, there is a m arked increase in the accum ulation o f storage reserves, cessation o f em bryogénie development, acquisition o f desiccation tolerance, and developm ent o f dorm ancy mechanisms (R ock and Q uatrano, 1995).
studies em ploying two main approaches: I )results from in vitro experim ents, and 2)resuits from correlation studies. However, both these approaches have limitations. Because culture conditions are very different from the situation in the developing seed environm ent, results from in vitro studies do not com pletely describe events in vivo. Hormone concentrations used in culture are generally several orders o f m agnitude higher than those found in seeds developing in vivo. The second approach relies on the
determination o f endogenous hormone levels in developing seeds followed by an attempt to correlate any changes in hormone levels with events occurring at various
developmental stages. The main problem with studies using this approach, apart from the limitations inherent in the m ethodologies used, is the assum ption that detection o f high hormone levels at a particular developmental stage means the hormone is responsible for the observed event. As Trewavas ( 1981) pointed out, the ability o f the tissue to respond to the horm one (tissue sensitivity) should also be taken into account. A change in tissue sensitivity could be a reflection o f a change in the level o f hormone receptor expression or changes in affinity by the receptor (Davies, 1995; Fim , 1986). So, while both these approaches are an invaluable first step in investigating the role played by hormones during seed development, the attention is shifting towards the use o f genetic and molecular studies to further understand the mechanisms involved in this process.
2.4.2.1 The role o f ABA during angiosperm seed development
Studies investigating the role o f plant hormones in the regulation o f seed development have centered on the role played by abscisic acid (ABA) and these have been largely confined to angiosperm species with short reproductive cycles. The com m on pattern o f A B A accumulation in a majority o f angiosperm seeds starts w ith low levels early in em bryo development, followed by an increases to peak levels in m id-to-late em bryogenesis, and finally a decline to low levels in mature seeds (Ackerson, 1984; Neill
et al., 1987; Finkeistein et al., 1985; G root et ai., 1991; Rock and Q uatrano, 1995). A biphasic A BA accum ulation pattern has been reported in a few species during embryo developm ent, including Phaseoliis coccineus (Perata et al., 1990).
A BA has both stimulatory and inhibitory roles at different tim es during seed developm ent. The hormone has been im plicated in the prevention o f precocious germ ination, w hich allows the em bryo to mature before germ ination can occur (A ckerson, 1984; Neill et al., 1987; Pence, 1991), in the acquisition o f desiccation tolerance (Pence, 1991; 1992; Ooms et al., 1993), in promoting storage protein synthesis (Oom s et al., 1993; Farrant et al., 1996; Finkeistein et al., 1985; Barratt and Clark. 1991; Berge et al., 1989; Kermode et al., 1989), and in the developm ent o f dorm ancy
m echanism s (Karssen et al., 1983; Hole et al., 1989; Ooms et al., 1993; Le Page-Degivry and G arello, 1992). Lipoxygenase activity during em bryo developm ent coincides with ABA synthesis and action and this has been suggested as a key enzym e catalyzing ABA synthesis from carotenoids (Belefant and Fong, 1991).
2.4.2.1.1 Precocious germination
Correlations between high endogenous ABA levels in immature em bryos and their failure to germinate have been reported in several species. During early
em bryogenesis o f soybean and Brassica napiis, ABA stimulates growth and storage protein accum ulation while inhibiting precocious germ ination (Ackerson, 1984; Finkeistein et al., 1985). Phaseoliis coccineus embryos showed a biphasic pattern of A BA accum ulation, the first peak coinciding with the time o f rapid em bryo grow th at the early cotyledonary stage (Perata et al., 1990). In alfalfa, while isolated em bryos at all stages o f developm ent germinate in vitro on culture m edia supplem ented w ith 3% sucrose, germ ination is slow er during mid-developm ent, the tim e o f highest endogenous A BA concentration (X u and Bewley, 1991). Le Page-Degivry and G arello, ( 1992)
reported that exposure o f young non-dorm ant sunflower embryos to exogenous ABA in culture also prevented precocious germination.
Further evidence com es from experim ents showing the occurrence o f precocious germ ination in the presence of reduced endogenous ABA levels. M id-developm ent alfalfa em bryos excised from seeds treated with the ABA biosynthetic inhibitor,
fluridone, have much higher germination rates than untreated em bryos (X u et al., 1990). Prem ature drying, w hich also depletes endogenous ABA in embryos and endosperm , results in precocious germ ination o f immature em bryos of soybean (Ackerson, 1984) and castor bean (Kermode et al., 1989).
A BA may act by inhibiting gibberellin’s effects in immature grass seeds e.g. barley. A switch from a developm ental mode to a germinative mode is paralleled by an increase in gibberellin-responsiveness, which can be examined by m easuring a-am ylase activity. The antagonistic effects of ABA and GA on a-amylase have been reponed during germination in barley aleurone (W ang et al., 1996). Treatments w hich lower ABA levels and permit precocious germination o f immature seeds, such as prem ature drying, induce a-am ylase production (C om ford et al., 1986).
Sensitivity to A B A appears to change w ith the developmental stage o f the em bryo. Kermode et al. (1989) reported that im posed drying of immature castor bean em bryos causes a 10-fold decrease in sensitivity o f isolated embryos to exogenous ABA in culture. Excised m ature em bryos that have undergone natural drying in the seed have a sim ilar response to exogenous ABA (Kermode et al., 1989). In Brassica napus
(Finkeistein et al., 1985) and alfalfa (Xu and Bewley, 1991), germ ination is suppressed by low er levels o f exogenous hormone for em bryos that have not begun to desiccate, while desiccating em bryos require higher concentrations. Further support com es from the observation that immatiure em bryos cultured prior to dehydration synthesize storage proteins in response to A B A while older im m ature embryos do not (Finkeistein et al,
It has been suggested that a reduction o f ABA in the em bryo and surrounding tissue alone is insufficient to evoke a switch to germination (K ermode et al., 1989). Support for this statem ent comes from several studies. Endogenous ABA levels decline in desiccating Brassica napus em bryos entering a period o f developm ental arrest, indicating that factors other than ABA are involved in blocking germ ination during the desiccating phase (Finkeistein et al, 1985). In addition, whole seeds from fluridone- treated alfalfa pods do not express vivipary i.e. do not precociously germ inate before maturity (Xu et al., 1990). In alfalfa, m id-to-late stage em bryos excised and placed on w ater germinate faster and with greater frequency than whole seeds at the same developm ental stages (Xu and Bewley, 1991). These results and the observation that isolation o f immature em bryos from whole seeds often results in germ ination (K ermode et al., 1989; Xu and Bewley, 1991; Xu et al., 1990) suggest that the surrounding seed environm ent may have an inhibitory effect on precocious germ ination.
This raises the question o f what additional factor present in the surrounding seed environm ent causes this inhibition o f germ ination? Xu et al.( I990)and Xu and Bewley (1991) reported that precocious germ ination o f excised im mature alfalfa em bryos can also be inhibited by application o f osm oticum to the culture medium. Furtherm ore, synthesis o f proteins unique to developing em bryos in situ only occurred in excised im mature em bryos cultured in the presence o f osmoticum, while ABA caused synthesis o f proteins which arise during maturation (Xu et al., 1990). A lthough ABA and
osm oticum both inhibit germination, the fact that fluridone-treated alfalfa seeds do not express vivipary suggests that osm oticum may be more im portant in maintaining em bryos in a developm ental mode in some species.
2.4.2.1.2 Storage reserve accumulation
ABA has been im plicated in the expression o f storage protein genes in developing seeds. M ost o f the evidence com es from studies which found a correlation between high
